Journey to Perfection

[keith.lamothe]

Any sort of character penalty that causes more time lost than glyph-swapping will discourage play of these awesome missions. 

Personally, all that argument does is make me want to remove glyph-swapping.....

That's basically my first reaction as well, but one lesson we've learned in developing this game is that any kind of really permanent heavy loss drives most of the audience (of this game) away, and either that bit of content (or the entire game, if it's a core thing) has to be a really niche-appeal thing or it has to be changed to not have that kind of loss potential.  This was disturbing to me since I like making high-consequences stuff, but since I'm not aware of an alternate sentient species with higher risk-tolerance to which to market the game...

If the game does well, then we can add more niche stuff to it, kind of like how hybrids and exos and other more-threatening stuff was added to AIW way post-release.

[x4000]

A few notes in response:

1. Glyph swapping isn't going anywhere.  That's hugely important to other aspects of the game, some of which haven't really come up yet.  Nor is that going to get less frequent or harder to do; so that's really not something that's up for debate.  This argument needs to be looked at in light of that part being set in stone.

2. When you get right down to it, anything is arbitrary.  Why are enemies one-shotting you and you one-shotting them?  It's arbitrarily done because it's fun and tense.  It's a different way to play.

3. What these missions are not meant to be is something that is only for the hardest of the hardcore.  Yes, I don't expect any but hardcore platformers to really enjoy lava escape, because they will lose a lot and losing isn't very fun.  But losing AND then getting a kick in the pants on top of it (or losing AND having to waste time going to town to swap a glyph) is punitive to the player.  It's very much not what this game is about, or what it has ever been about, as the core experience.

4. If these were just one-off experiences that weren't any fun for the non-hardest-core, that would be one thing.  But I'm the sort, come to that, who plays on Hero rather than Chosen One, but I really love the journey to perfection and lava escape missions both.  Talk about tense!  Thing is, if I lose, I just want to have lost an opportunity, not actively pushed backwards in the rest of my progress.

Really what it boils down to is suddenness.  If I am on a mission or any sort, or an expedition, and things are going poorly, then I can flee to town.  I can give up and go away.  If I don't do so, then that's my own fault if my character dies.  It's a choice, and an interesting one.  But for lava escape and journey to perfection, what they mostly do is make the bulk of people avoid them with characters they care at all about.  The missions are fun, but there is no way to sense that something is going poorly and then retreat or push on.  That's fundamentally at odds with the rest of the game.

Which, from a gameplay standpoint, is great: we want varied modes of accomplishing goals, and the more fun ways of doing missions there are, the better.  But if some come with insane extra risks for no additional reward, that's a problem.

But you're right there is a problem with consistency.  The unconsciousness thing bugs me, too, even though I was halfway through implementing it.  I'm going to do something else instead -- basically make it so that you can choose to play as an avatar instead of as yourself on these missions.

[x4000]

Terraziel, turns out you did convince me.  We need the death to always be permanent, period, and we also need to not introduce something with the complexity and hokiness of an avatar.

Really the problem is the hassle of switching between characters in your settlement, and the travel time of running back there and then to the mission again.  But the ability to swap out characters at will is still ripe for abuse.  So I'm still thinking.

Here's the goals that I, personally, have for whatever gets done here.  These I'm pretty much unwilling to compromise on, but varying solutions that meet all the criteria I'm open to.

1. Death must remain permanent.  When you are controlling a character and that character dies, they remain dead.  Period.  No unconsciousness, no "extra lives," no take-backs of any kind.  No way to control a remote robot that does the dangerous work for you.  Period.  This is central to the game.

2. The ability to switch between characters is equally important.  Building up a stable of varied characters is something I want to encourage, and will play into later strategic decisions further down the line.  It may seem trivial now, but longer-term this is going to be a central mechanic of the game.

3. We want to maintain a variety of gameplay styles, including some of which are naturally going to be more niche.  However, we don't want to needlessly exclusionary with those modes.  The JTP and LE missions are both fun for certain types of personalities, and we want for people with those personalities to enjoy them.  But having a lot of the playerbase scared off of even trying them is not good.  That was why we made the mark 1 of all the turrets and starships free to players at the start of AI War: players wouldn't take a risk on something unknown if it was perceived as having too high of a risk or opportunity cost.

4. Making the player run around a huge amount in order to "play optimally" but at the expense of their own time is bad bad bad.  This is like why knowledge raiding was such as bad thing in AI War and we had to change it to be more hardcore and interesting.  And Penumbra is absolutely right that, as it stands, JTP and LE missions at the moment fall prey to this.  Because they are fun, and the optimal way to play them is to go and swap the glyph to some nobody and then come back and do this mission.  Which is the core conundrum, and what is broken about the current implementation at the moment.

So right now, #1 and #2 are being handled properly.  #3 is being handled properly except that we're scaring some people off.  #4 is really badly mishandled right now with these two mission types, and I'm not sure yet what to do about it.

[Terraziel]

Alright, as far as number 4 is concerned, if we accept that switching characters is the optimal way to do it, then the obvious solution is to just run with it.

Set up a "character swap" terminal inside the mission prep area, so essentially a long range glyph transfer, and just let people do it, make it take a glyph transfer scroll as "ammo" if necessary, if you limit it to the mission prep area for these missions then it is not particularly abusable at the very least it is probably easier to actually go back to your settlement rather than seek out a particular mission type. I mean arguably this removes the opportunity for the "new" character to die getting to the mission but given that you only ever need to traverse a single chunk there and back to the settlement it's not really a major problem.

For number 3, how about applying a similar logic to the Vortex missions? so initially you can take 3 hits in a Journey mission, but after completing a few it goes down to two, then 1.
For Lava Escape it could be that simply the speed of the lava increases, or maybe the complexity of the level increases, so that the initial missions are more or less a single straight line, then more branches and dead ends are added as you complete them.

This would ease players into the missions, on Journey missions allowing them to retreat if they think they are in over their heads, for lava missions....well it would at least help them figure out if they like that sort of challenge, but I can't think of a good way to allow players an out in them, maybe simply providing an elusion scroll in the initial ones.

[x4000]

These are all really good points.

Keith, what kind of havok would we be talking about for a "switch character" terminal of that sort?  It basically brings up the existing View NPCs buttons that you already have in the planning menu, but then has an extra button which is "swap characters" or something.

We'd have to switch some things about the NPC records, and physically move both of the characters into and out of their respective chunks on the server and then alert the various clients on both sides.  For the player instance data itself, not much would change except the entity it is controlling, which we already do when somebody dies or whatever else.

Thoughts?

[Martyn van Buren]

Maybe I'm too late to convince anybody, but I really think you want something along the lines of unconsciousness at least for lava escape missions, and maybe also for journey, although I haven't tried it yet.  I also want to argue that you don't want missions to disappear when you fail them, although that perhaps could be something affected by strategic difficulty.

With the greatest respect, I think you guys are kind of misunderstanding this issue because you come at it from the perspective of really highly-skilled platform gamers.  It seems like you're able to play true platforming games and seriously expect losing lives to be a rare event.  For most players --- I think I fall somewhere in the range of the average --- platforming is a really hard skill and the way you play a lot of the time is throwing a lot of lives at learning a level and getting past it.  I think ruling out this kind of play will effectively gate off missions with instant-death situations from most players --- basically, it seems to me that it says if you don't know how to beat the mission already, you shouldn't try to practice and get better.

I'd be very happy with a solution that replaces only the instant-death part of the mission with unconsciousness, leaving other traps and ways to take real damage.  I'd also be happy with leaving lava escapes as they are now and creating a kiddie version for players like me, which I suggested some time ago.  Or anything else, really, that wouldn't make me feel like such a chump for wanting to try a mission type I don't have any talent for.

Of course, glyph transfering and using a lot of new characters remains a way around the hard cost of losing a character, but at a significant cost in terms of the atmosphere of the game.  I've done it and won a lava escape, but after learning everywhere else in the game that I'm supposed to protect my characters I feel like I'm doing something horribly wrong when I wind up back on the selection screen a dozen times in five minutes.

[keith.lamothe]

Keith, what kind of havok would we be talking about for a "switch character" terminal of that sort?  It basically brings up the existing View NPCs buttons that you already have in the planning menu, but then has an extra button which is "swap characters" or something.

We'd have to switch some things about the NPC records, and physically move both of the characters into and out of their respective chunks on the server and then alert the various clients on both sides.  For the player instance data itself, not much would change except the entity it is controlling, which we already do when somebody dies or whatever else.

It would mean generalizing glyph-transplant so that the target npc doesn't have to be in the same chunk (or any loaded chunk) which is a significant amount of havoc at this point but not excessive.  It's just the server/singleplayer processing of it, since the multiplayer-client processing of it already doesn't require that either the using or the target entity's chunk be loaded because it has to be processed for everyone in order for them to be up-to-date on who is running with who.

Other than that, it's just a matter of making this entity which pulls up the view-npc interface and a bool for that where clicking on an npc literally triggers the GlyphTransplant ability (or, if you prefer, a clone of said that is identical except it requires no item).

[x4000]

Martyn van Buren: It is a very good point that practice is the road to mastery, and anything which makes practice impossible makes a mission infeasible for anyone who isn't already a master.

And you're right that Terraziel and I are both terrible ones to design a solution here.  He plays on Chosen One, for cripes sake, and is clearly far more skilled at the combat side of this than I.  I'm not a perfect platform gamer, but I lose maybe three lives in my average playthrough of a new 2D Mario game on my first try.  So I come at it from the perspective of wanting it to be a fun and interesting challenge the first time, because the first time is all there is.

With most of the game, these are not at odds.  The difficulty level allows for levels of granularity that let us all have an experience that matches our own individual blend of skills, learn, and get better.  These two mission types tend to preclude learning because they are so all-or-nothing, though.

When the currency you are playing with is your life, in this sort of fashion, that's really kind of problematic.  Having no margin for error is great for existing masters, it's fun and tense and interesting.  But these really do preclude anyone but existing masters learning them.

Perhaps that's the way it should be, I guess: maybe these are just those optional hardcore side jaunts that nobody but the top 10% should even bother trying to play.  But a big part of me thinks that this really is a fun sort of mission to play, and would be interesting to more than 10% of our audience.  So gating it off behind extreme difficulty is basically just "stepping in it" the same way that nightmare mode did in that other game.

Bleh.

[x4000]

Keith, what kind of havok would we be talking about for a "switch character" terminal of that sort?  It basically brings up the existing View NPCs buttons that you already have in the planning menu, but then has an extra button which is "swap characters" or something.

We'd have to switch some things about the NPC records, and physically move both of the characters into and out of their respective chunks on the server and then alert the various clients on both sides.  For the player instance data itself, not much would change except the entity it is controlling, which we already do when somebody dies or whatever else.

It would mean generalizing glyph-transplant so that the target npc doesn't have to be in the same chunk (or any loaded chunk) which is a significant amount of havoc at this point but not excessive.  It's just the server/singleplayer processing of it, since the multiplayer-client processing of it already doesn't require that either the using or the target entity's chunk be loaded because it has to be processed for everyone in order for them to be up-to-date on who is running with who.

Other than that, it's just a matter of making this entity which pulls up the view-npc interface and a bool for that where clicking on an npc literally triggers the GlyphTransplant ability (or, if you prefer, a clone of said that is identical except it requires no item).

Yeah, and we're past the point where we can do something like that prior to 1.0.

[Terraziel]

I'd also be happy with leaving lava escapes as they are now and creating a kiddie version for players like me, which I suggested some time ago.  Or anything else, really, that wouldn't make me feel like such a chump for wanting to try a mission type I don't have any talent for.

I think creating the separate mission type is probably the best solution, or at least strikes me as the way most likely to please everybody. The lava escape missions are sort of the far end of the platforming scale, putting something in the middle ground so that people can get a similar experience, and then choose whether or not they can or wish to go up to the lava version seems fair to me.

With the Journey missions, though obviously as noted I am a poor example, I find these to be less about skill and more about planning, "instant death" sounds bad but as long as you proceed slowly and carefully I think anyone should be able to do them on their chosen difficulty. If my previous suggestion for easing people into them is implemented all the more so.

[tigersfan]

Martyn van Buren: It is a very good point that practice is the road to mastery, and anything which makes practice impossible makes a mission infeasible for anyone who isn't already a master.

And you're right that Terraziel and I are both terrible ones to design a solution here.  He plays on Chosen One, for cripes sake, and is clearly far more skilled at the combat side of this than I.  I'm not a perfect platform gamer, but I lose maybe three lives in my average playthrough of a new 2D Mario game on my first try.  So I come at it from the perspective of wanting it to be a fun and interesting challenge the first time, because the first time is all there is.

With most of the game, these are not at odds.  The difficulty level allows for levels of granularity that let us all have an experience that matches our own individual blend of skills, learn, and get better.  These two mission types tend to preclude learning because they are so all-or-nothing, though.

When the currency you are playing with is your life, in this sort of fashion, that's really kind of problematic.  Having no margin for error is great for existing masters, it's fun and tense and interesting.  But these really do preclude anyone but existing masters learning them.

Perhaps that's the way it should be, I guess: maybe these are just those optional hardcore side jaunts that nobody but the top 10% should even bother trying to play.  But a big part of me thinks that this really is a fun sort of mission to play, and would be interesting to more than 10% of our audience.  So gating it off behind extreme difficulty is basically just "stepping in it" the same way that nightmare mode did in that other game.

Bleh.

What about making it based on difficulty level? I know for me, I'm more with MvB on this one. I've never completed an entire Mario game, so I'm not a great platformer, but the idea of these missions does sound fun, and I'd like to play them some time, but I'm not likely to because the chance of failure is so high. So, what about if I set the platforming level low, I wouldn't necessarily lose a life, and maybe just get knocked out. But someone on "I'm already the guy", if he were to fail loses the character AND maybe even loses the mission and can no longer attempt it.

[Bluddy]

I think the problem here is that the glyph transfer concept as it is is not very compatible with the mission concepts from GTA (which is what we have). There's a reason GTA3 was the first one to really nail the formula perfectly: making death something you recover from instantly (albeit without weapons) allows you to try crazy-hard missions (among other things), learn from your mistakes, try different approaches etc. The glyph model + character specific bonuses is part of a very different model, suited more to the parts where you build up resources and eventually take down the overlord: you have to think tactically, constantly consider the possibility of defeat, retreat when the odds are against you etc. The goal is to synthesize these 2 models, and I don't know if you can set too many preconditions for that synthesis.

If we try to be more precise, the real problem here is the character bonuses from stones and losing them when we lose a life. Losing character bonuses is a drag because suddenly you're forced to do a lot of grinding. And if you think about it, do we really want that to be the threat? Yes, it's something that makes you tense -- you don't want to lose this character you've built up because you'd have to grind all these houses. But is that a good thing? Maybe that's not a good threat. The threat of death should perhaps come from somewhere else.

So here's a slightly different model. It doesn't necessarily obey the points, but I don't know if that's a bad thing. Your glyph can only inhabit each of the characters that are currently in your settlement. If a character dies, he's mortally wounded and you can't occupy his body, but the Ilari are slowly healing him. Alternatively he's dead but his soul gets reincarnated/inserted into a wanderer in the settlement and that takes time (but allows for the vengeful ghost concept + no unconsciousness). But his upgrade points aren't lost -- that's a threat that just translates to grind. But what happens once all of your bodies (which are now the equivalent of 'lives') are killed and awaiting reincarnation? Your settlement is defenseless and either you get luck and a wanderer strays into your settlement  (a random event that's unlikely), or a lot of time passes by without the ability to do anything. Any missions you wanted are now gone, some CP has been gained without acquiring any precious resources (enemy tier might become higher), and in the post 1.0 game as I imagine it, the enemy will move without you being able to respond.

How does this model handle the missions at hand? The cost of losing one life isn't so great since the modifiers you had for that character will come back once that character reincarnates. So the cost is just a small bit of CP, which is not too bad. But if it's the last living guy in your settlement, the cost will be much higher, so you'll be much more careful.

This also suggest for a catastrophic loss of a survivor as a serious random/punishment thing. That's equivalent to taking away a life.

[BobTheJanitor]

Re: Glyph swapping terminals.

I know this is being thought of purely from a gameplay perspective right now, but give a little consideration to the story and lore side of things here. I, the brave glyphbearer risked life and limb to explore the fractured world and rescue some survivor huddled in a dark basement full of rampaging beasts, then took them back to a settlement where they can be protected and nurtured... Until I need an expendable body to send into extreme danger, then I just call them up against their will to run from rising lava or fight monsters so strong that one touch is instant death. Meanwhile I'll be back in the settlement sipping tea and waiting to hear word of their demise.

It... doesn't exactly feel heroic, does it? And yes, the base mechanic of swapping for high risk missions already has this implication as well, but having a little terminal to just teleport them in and take them over really sort of brings it to the forefront.

[Terraziel]

What about making it based on difficulty level? I know for me, I'm more with MvB on this one. I've never completed an entire Mario game, so I'm not a great platformer, but the idea of these missions does sound fun, and I'd like to play them some time, but I'm not likely to because the chance of failure is so high. So, what about if I set the platforming level low, I wouldn't necessarily lose a life, and maybe just get knocked out. But someone on "I'm already the guy", if he were to fail loses the character AND maybe even loses the mission and can no longer attempt it.

I'd thought about this and the concern I had was that it wasn't granular enough, I mean, not to sound silly but by lowering the difficulty... you'd lower the difficulty not just the risk, and then if someone finds the lava too slow they have no choice but to up the "risk" as well, getting the balance right would be complicated at best.

Re: Glyph swapping terminals.

I know this is being thought of purely from a gameplay perspective right now, but give a little consideration to the story and lore side of things here. I, the brave glyphbearer risked life and limb to explore the fractured world and rescue some survivor huddled in a dark basement full of rampaging beasts, then took them back to a settlement where they can be protected and nurtured... Until I need an expendable body to send into extreme danger, then I just call them up against their will to run from rising lava or fight monsters so strong that one touch is instant death. Meanwhile I'll be back in the settlement sipping tea and waiting to hear word of their demise.

It... doesn't exactly feel heroic, does it? And yes, the base mechanic of swapping for high risk missions already has this implication as well, but having a little terminal to just teleport them in and take them over really sort of brings it to the forefront.

Lets be clear I suggested it and I don't like it for much the same reasons, but If it's the price to pay, one shall pay it, I just won't use them. I suppose it's arguable that it's actually pretty sensible from the settlements perspective, who do you send on Operation: Certain Death? cannon fodder, not your elite troops. Not very heroic obviously but perhaps indicative of the moral collapse caused by the apocalypse.

[x4000]

Terraziel: I agree on the need for some sort of middle of the road mission, as well as the fact that difficulty isn't granular enough.  However, I also think that completing a lot of the mission in question then making it ALWAYS harder is also a bad thing.  "Congratulations, now you've done a lot of something you've liked, so now we've turned it into something that has a lot more risk and that you therefore don't like!" 

I think what we really need to do is make two new mission types: lava escape lite, and journey to something-less-than-perfection.  Completing these two a few times would then unlock the ultra-hardcore versions that the hardcore could complete, basically with more risk but otherwise the same general idea of gameplay.  Nobody is missing out on anything, and we keep the same general model.

Bluddy: You make some good points, as always, but in this case there's just no way we could switch up the model to that degree and have any sort of a sane 1.0 version.  And honestly I don't think that particular sort of change would be very positive, because it would actually increase the penalty for death (settlements getting low on settlers, etc), the CP ever increasing without your getting tier orbs would fundamentally break the game (making it possible to lose), and various other problems.  That's a whole other game you're basically describing, and it would take months to get there. 

Given that you have to go explore some houses and other buildings every so often anyhow, just looking for all sorts of things, the loss of upgrade points isn't normally a giant deal in the first place unless you are really sending guys through the grinder.  Which, as you pointed out, the game is encouraging you not to do as much as it can.  Which really does make these two missions kind of incompatible with the rest of the game, as you pointed out.

BobTheJanitor: Agreed on the thematic issues, but as you alluded to those are somewhat secondary.


TLDR: I think we need to think of new, less-hardcore rules for these types of missions.  And then make those the ones that get unlocked early, and these hardcore versions unlocked as an optional alternative by winning those ones a bit.

--------------

So we've got something like "Water Escape" as martyn_van_buren's lava escape lite: http://www.arcengames.com/mantisbt/view.php?id=6318

It gives you an Acid Gills instead of a heatsuit, and so the water doesn't harm you, but it does majorly slow you down and make things harder.  If the water reaches the exit at the top of the level before you, then you lose the mission but nothing bad happens to you personally.

So how would it work for "Journey To Perfection" lite?  One idea: you're guiding some sort of "flying buddy" to the exit.  AA missiles don't notice him, but they do notice you.  Your damage against enemies is increased, but the damage you take is not.  However, one hit will kill your buddy, and he sticks really closely to you.  If he dies, you lose.

Therefore, in both missions you still retain the usual risk of death: get hit by a lot of monsters and you die.  But on the other hand, there's not any form of instant death condition in either.  For that matter, this could become the SOLE method of journey to perfection, in my opinion.  But lava escape is still interesting outside of watery escape, potentially.

--------------

Or maybe both of these just need to be plain-out altered.  Heatsuits make you invulnerable to lava, but don't aid your movement in it.  This would be helpful for a number of places, where falling in the lava isn't therefore going to be death-in-5-seconds, but rather making you super easy prey for monsters.  That wouldn't really make the lava flats any easier, for instance, but it would turn the existing lava escape into the proposed watery escape pretty much in one go.

And then just changing JTP to be the "guard your buddy" type of mission where he's at one-shot risk and you're not, and we've got ourselves another mission that is more in keeping with the general spirit of the game, while still retaining the hard challenge of actually completing these things.

The stakes are lower, making things less tense in a lot of respects, but that's... well, that's in keeping with the rest of the game, too.  "Death is supposed to be the culmination of a series of mistakes," as I keep saying over and over again.  And right now even the existence of lava without some defense (heatsuits not protecting you at all) is counter to that.  Heatsuits could even just reduce the damage from lava by something like 100x, so instead of 5 seconds to escape you have 500 seconds, which is 8.3 minutes.  If you can't get out in 8 minutes, then you pretty much have it coming to you.   That would keep the tenseness of "lava always damages me," while making it so that lava isn't such an instantly-deadly threat.

Thoughts?